Picture for Tomotake Sasaki

Tomotake Sasaki

D3: Data Diversity Design for Systematic Generalization in Visual Question Answering

Add code
Sep 15, 2023
Viaarxiv icon

Modularity Trumps Invariance for Compositional Robustness

Add code
Jun 15, 2023
Viaarxiv icon

HICO-DET-SG and V-COCO-SG: New Data Splits to Evaluate Systematic Generalization in Human-Object Interaction Detection

Add code
May 17, 2023
Viaarxiv icon

Deephys: Deep Electrophysiology, Debugging Neural Networks under Distribution Shifts

Add code
Mar 17, 2023
Viaarxiv icon

Safe Exploration Method for Reinforcement Learning under Existence of Disturbance

Add code
Sep 30, 2022
Figure 1 for Safe Exploration Method for Reinforcement Learning under Existence of Disturbance
Figure 2 for Safe Exploration Method for Reinforcement Learning under Existence of Disturbance
Figure 3 for Safe Exploration Method for Reinforcement Learning under Existence of Disturbance
Figure 4 for Safe Exploration Method for Reinforcement Learning under Existence of Disturbance
Viaarxiv icon

Transformer Module Networks for Systematic Generalization in Visual Question Answering

Add code
Jan 27, 2022
Figure 1 for Transformer Module Networks for Systematic Generalization in Visual Question Answering
Figure 2 for Transformer Module Networks for Systematic Generalization in Visual Question Answering
Figure 3 for Transformer Module Networks for Systematic Generalization in Visual Question Answering
Figure 4 for Transformer Module Networks for Systematic Generalization in Visual Question Answering
Viaarxiv icon

Do Neural Networks for Segmentation Understand Insideness?

Add code
Jan 25, 2022
Viaarxiv icon

Symmetry Perception by Deep Networks: Inadequacy of Feed-Forward Architectures and Improvements with Recurrent Connections

Add code
Dec 08, 2021
Figure 1 for Symmetry Perception by Deep Networks: Inadequacy of Feed-Forward Architectures and Improvements with Recurrent Connections
Figure 2 for Symmetry Perception by Deep Networks: Inadequacy of Feed-Forward Architectures and Improvements with Recurrent Connections
Figure 3 for Symmetry Perception by Deep Networks: Inadequacy of Feed-Forward Architectures and Improvements with Recurrent Connections
Figure 4 for Symmetry Perception by Deep Networks: Inadequacy of Feed-Forward Architectures and Improvements with Recurrent Connections
Viaarxiv icon

Three approaches to facilitate DNN generalization to objects in out-of-distribution orientations and illuminations: late-stopping, tuning batch normalization and invariance loss

Add code
Oct 30, 2021
Figure 1 for Three approaches to facilitate DNN generalization to objects in out-of-distribution orientations and illuminations: late-stopping, tuning batch normalization and invariance loss
Figure 2 for Three approaches to facilitate DNN generalization to objects in out-of-distribution orientations and illuminations: late-stopping, tuning batch normalization and invariance loss
Figure 3 for Three approaches to facilitate DNN generalization to objects in out-of-distribution orientations and illuminations: late-stopping, tuning batch normalization and invariance loss
Figure 4 for Three approaches to facilitate DNN generalization to objects in out-of-distribution orientations and illuminations: late-stopping, tuning batch normalization and invariance loss
Viaarxiv icon

Annotation Cost Reduction of Stream-based Active Learning by Automated Weak Labeling using a Robot Arm

Add code
Oct 03, 2021
Figure 1 for Annotation Cost Reduction of Stream-based Active Learning by Automated Weak Labeling using a Robot Arm
Figure 2 for Annotation Cost Reduction of Stream-based Active Learning by Automated Weak Labeling using a Robot Arm
Figure 3 for Annotation Cost Reduction of Stream-based Active Learning by Automated Weak Labeling using a Robot Arm
Figure 4 for Annotation Cost Reduction of Stream-based Active Learning by Automated Weak Labeling using a Robot Arm
Viaarxiv icon